Increased risk of fractures among people with type 1 and type 2 diabetes in the population of Finland
Tomi Istolainena, Heikki Krögera, Reijo Sunda
a) Kuopio Musculoskeletal Research Unit, Institute of Clinical Medicine, University of Eastern Finland, Kuopio Finland
b) Orthopaedic, Traumatology and Hand Surgery Clinic, Kuopio University Hospital
Abstract
Aims/hypothesis
Diabetes mellitus (DM) is a worldwide disease affecting hundreds of millions of patients. Studies suggest an association between diabetes and risk of fractures. The purpose of this study was to compare the relative fracture risk in subjects with and without diabetes. The second aim was to compare the relative fracture risk among women and men with type 1 and type 2 diabetes.
Methods
This is a population-based cohort study of the population of Finland. To identify people with diabetes and fractures, we used the register data collected for the FinDM (Diabetes in Finland) project. The fractures in the general population were extracted from the Care register for Health Care (CRHR). In these two data sets, all admissions with any fracture diagnosis were identified from the CRHR for the years 1998–2017. The fracture incidences were stratified by age and sex.
Standardised incidence ratios (SIRs) were derived.
Results
DM was associated with an increased risk of all fractures, especially distal radius, proximal humerus, vertebral, and hip fractures. Low-energy fractures were more common among subjects with DM under the age of 85 (SIR 1.07–2.08). High-energy fractures were more common among subjects with DM under the age of 60 in men (SIR 1.02–1.43) and under the age of 50 in women (SIR 1.08–1.48). Subjects with Type 1 DM had a higher fracture risk compared to those with type 2 DM.
Conclusions
DM was associated with an increased risk of all fractures in both genders. People with type 1 diabetes had higher relative fracture risk compared to people with type 2 diabetes. This study showed that, in a large proportion of fractures, the relative fracture risk associated with diabetes is noticeable, especially in adolescents. The high-energy fracture sensitivity of people with diabetes needs to be investigated further.
Introduction
Diabetes mellitus is a worldwide disease affecting hundreds of millions of patients. Studies suggest that diabetes is associated with an increased risk of fractures(1–6), but the exact mechanism which predisposes patients with diabetes to fractures is not clear. Patients with long-term disease, poor glycaemic control, β-cell failure, and who receive insulin treatment are at the highest risk of fractures (3,5,7,8). There is evidence that, in diabetes, bone regeneration is slowed, leading to microstructural changes and compromised bone mineral properties(9).
Increased risk of vertebral and non-vertebral fractures, especially in the hip, forearm and humerus, among people with diabetes has been reported(7,10–14). The risk of fractures and osteoporosis is increased, especially in patients with type 1 diabetes(3,4,11,12). Among people with type 2
diabetes, bone mineral density can be reduced, normal or increased(15,16); however, recent studies suggest that it is also associated with an increased risk of fractures (6,9,13). The incidence of vertebral, wrist, shoulder, elbow, ankle and hip fractures is increased among patients with type 2 diabetes compared to people without diabetes(6,17–19).
Our study used extensive data to compare people with and without diabetes, as well as type 1 diabetics and type 2 diabetics, by age group and fracture type using standardised incidence ratios.
Our study used the population of Finland, over 5 million subjects, followed from 1998 to 2017. In previous studies, the number of study subjects and fractures has been relatively small. For example, a population-based 6-year follow-up study was conducted, following 27,159 subjects from the municipality of Tromsø from 1994 until 2001 (12). Also, there are no large studies comparing low and high-energy fractures among people with type 1 and type 2 diabetes.
The purpose of this study is to examine standardised incidence ratios for each type of fracture in women and men of all ages with diabetes in Finland between 1998 and 2017, using comprehensive register data, and to compare this to people without diabetes. We also separately determined the standardised incidence ratio for women and men with type 1 and type 2 diabetes by fracture type.
Material and methods
Data from two separate research projects were used as the study material. The primary source of data was the FinDM (Diabetes in Finland) project, which contains all diagnosed Finnish diabetes cases with their healthcare histories since 1964 from the KELA (the Social Insurance Institution of Finland) special reimbursement register, the prescription register, the hospital discharge register and the causes-of-death register(20). The second dataset originated from the Fractures in Finland
project, which was used to calculate the numbers of fractures in the general population based on the data extracted from the Care Register for Health Care (CRHR). In these two datasets, all admissions with any fracture diagnosis were identified from the CRHR for the years 1998 to 2017. Fractures were classified using the International classifications of Diseases (Table 5), and further divided into four different fracture mechanisms based on the external cause diagnosis: low-energy, high-energy, unclear and external cause not recorded in the register. In cases of multiple treatment periods for the same patient and fracture type, the first admissions in each fracture category for each individual during the study period were considered to represent new fractures(21).
These data enable the risk of fracture to be calculated among patients with and without diabetes.
Patients were divided into groups by sex and age (5-year age groups, under 30 years and over 85 years). Data were available for the years 1998 to 2017.
Dividing the number of fractures by the corresponding time in person-years enabled us to calculate the incidence of fractures per 100,000 person-years for people with and without diabetes. The follow-up time of people with diabetes in the age and sex groups was calculated based on the FinDM data, accumulating a total of 6.08 million person-years during the follow-up period. The follow-up time in person-years for whole population was calculated based on average population figures obtained from the statistics and indicator bank Sotkanet(22).
To determine the differences in fracture risk between people with and without diabetes, we calculated standardised incidence ratios (SIR) for different age groups and for males and females separately. By multiplying the incidence of fractures among people without diabetes by the follow- up time among people with diabetes, the expected numbers of fractures were obtained, which describes how many fractures people with diabetes would be expected to have had if the incidence had been similar to that among people without diabetes. By comparing the observed and expected numbers of fractures, standardised incidence ratios reflecting the risk difference between people with and without diabetes were calculated by sex and age groups. Also, 95% confidence intervals (95% CI) were calculated for these SIR numbers.
Table 1. Standardised incidence ratio (SIR) of fractures by age groups in men and women with diabetes across 20 years (1998-2017)
Distal forearm Proximal Humerus Vertebral Hip Age groups Male
SIR(95% CL)
Female SIR(95%CL)
Male SIR(95%CL)
Female SIR(95% CL)
Male SIR(95%CL)
Female SIR(95% CL)
Male SIR(95%CL)
Female SIR(95% CL)
<30 1,22 (1,10-1,35)
1,22 (1,06-1,40)
1,61 (1,27-2,02)
2,02 (1,60-2,53)
2,05 (1,60-2,58)
2,62 (1,95-3,45)
1,77 (1,09-2,69)
3,49 (2,00-5,67)
30-34 1,62
(1,22-2,10) 1,36 (1,00-1,80)
3,04 (2,02-4,29)
3,98 (2,58-5,87)
1,44 (0,93-2,12)
1,65 (0,88-2,82)
35-39 1,30
(1,00-1,65) 1,35 (1,06-1,69)
2,67 (1,95-3,57)
2,37 (1,56-3,45)
1,81 (1,34-2,40)
1,82 (1,10-2,84)
30-39 2,72
(1,92-3,75)
3,86 (2,29-6,10)
40-44 1,33
(1,10-1,59) 1,20 (0,99-1,45)
2,56 (2,03-3,17)
2,64 (1,98-3,44)
1,63 (1,28-2,04)
1,79 (1,17-2,62)
2,62 (1,96-3,42)
4,00 (2,59-5,91)
45-49 1,28
(1,10-1,47) 1,07 (0,92-1,24)
2,38 (2,03-3,17)
2,62 (2,16-3,16)
1,57 (1,31-1,87)
1,57 (1,13-2,13)
2,22 (1,83-2,68)
3,29 (2,46-4,30)
50-54 1,07
(0,95-1,20) 0,89 (0,80-0,99)
1,75 (1,53-1,99)
2,10 (1,85-2,38)
1,52 (1,33-1,73)
1,37 (1,07-1,72)
2,09 (1,83-2.39)
2,09 (1,69-2,55)
55-59 0,99
(0,89-1,09) 0,72 (0,66-0,78)
1,57 (1,40-1,76)
1,81 (1,64-1,99)
1,45 (1,29-1,62)
1,62 (1,38-1,88)
1,49 (1,34-1,66)
2,01 (1,76-2,28)
60-64 0,86
(0,78-0,95) 0,72 (0,67-0,77)
1,43 (1,29-1,58)
1,80 (1,67-1,94)
1,12 (1,00-1,24)
1,52 (1,37-1,67)
1,27 (1,16-1,38)
1,87 (1,70-2,05)
65-69 0,90
(0,82-0,99) 0,70 (0,66-0,74)
1,22 (1,10-1,35)
1,63 (1,53-1,74)
1,21 (1,10-1,33)
1,51 (1,37-1,67)
1,26 (1,17-1,35)
1,54 (1,44-1,65)
70-74 0,87
(0,78-0,97) 0,69 (0,65-0,73)
1,29 (1,16-1.43)
1,40 (1,32-1,49)
1,35 (1,24-1,48)
1,33 (1,22-1,44)
1,13 (1,06-1,20)
1,34 (1,27-1,41)
75-79 0,72
(0,63-0,81) 0,71 (0,67-0,74)
1,06 (0,94-1,18)
1,24 (1,17-1,30)
1,22 (1,11-1,33)
1,22 (1,15-1,30)
1,09 (1,03-1,14)
1,20 (1,16-1,25)
80-84 0,92
(O,80-1,05) 0,72 (0,68-0,76)
1,07 (0,95-1,22)
1,15 (1,09-1,22)
1,17 (1,06-1,28)
1,11 (1,04-1,18)
0,96 (0,91-1,01)
1,07 (1,04-1,11)
>85 0,79 (0,66-0,93)
0,80 (0,76-0,85)
0,92 (0,79-1,07)
0,98 (0,93-1,04)
1,11 (1,00-1,23)
1,06 (1,00-1,12)
0,85 (0,81-0,89)
1,00 (0,97-1,03)
Table 2. Standardised incidence ratio (SIR) of low- and high-energy fractures by age groups in men and women with diabetes across 20 years (1998-2017)
Low-energy fracture
High-energy fracture
Age groups Male
SIR(95%CL)
Female SIR(95% CL)
Male
SIR(95% CL) Female
SIR(95% CL)
<30 1,38(1,29-1,48) 1,60(1,46-1,75) 1,43(1,32-1.55) 1,48(1,31-1,66)
30-34 1,96(1.73-2.21) 2,02(1.73-2.34) 1,42(1,20-1,66) 1,33(0,98-1,75)
35-39 1,94(1.76-2.15) 2,08(1.85-2.33) 1,43(1,25-1,64) 1,11(0,84-1,45)
40-44 1,99(1.84-2.14) 1,74(1.58-1.92) 1,25(1,12-1,40) 1,46(1,20-1,76)
45-49 1,71(1.61-1.81) 1,69(1.57-1.82) 1,23(1,13-1.34) 1,08(0,89-1,29)
50-54 1,60(1.53-1.68) 1,57(1.49-1.66) 1,02(0,95-1,10) 0,88(0,75-1,02)
55-59 1,40(1.35-1.46) 1,40(1.35-1.46) 1,02(0,96-1,09) 0,87(0,77-0,99)
60-64 1,31(1.27-1.36) 1,37(1.33-1.42) 0,84(0,79-0,90) 0,80(0,71-0,90)
65-69 1,28(1.24-1-32) 1,31(1.27-1.34) 0,87(0,81-0,93) 0,89(0,81-0,98)
70-74 1,29(1.25-1.34) 1,22(1.19-1.25) 0,86(0,79-0,93) 0,90(0,82-0,99)
75-79 1,19(1.15-1.23) 1,16(1.14-1.19) 0,84(0,77-0,92) 0,79(0,72-0,86)
80-84 1,10(1.06-1.13) 1,07(1.05-1.09) 0,90(0,81-1,01) 0,81(0,72-0,90)
>85 0,96(0,92-0,99) 0,99(0,98-1.01) 0,85(0,72-0,99) 0,86(0,76-0,97)
Table 3. Standardised incidence ratio (SIR) of fractures by age groups in men and women with type 1 diabetes compared to those with type 2 diabetes across 20 years (1998-2017)
Distal forearm Proximal Humerus Vertebral Hip Age groups Male
SIR(95% CL)
Female SIR(95%CL)
Male SIR(95%CL)
Female SIR(95% CL)
Male SIR(95%CL)
Female SIR(95% CL)
Male SIR(95%CL)
Female SIR(95% CL)
<30 2,08 (1,87-2,31)
2,85 (2,47-3,28)
1,22 (0,95-1,54)
1,36 (1,06-1,72)
0,96 (0,74-1,22)
1,18 (0,86-1,58)
0,53 (0,31-0,84)
0,47 (0,24-0,83)
30-34 0,95
(0,68-1,29) 1,81 (1,26-2,50)
2,28 (1,46-3,39)
1,28 (0,75-2,05)
1,99 (1,23-3,04)
2,01 (0,96-2,82)
35-39 1,71
(1,25-2,28) 1,87 (1,39-2,46)
0,98 (0,63-1,44)
0,91 (0,50-1,52)
0,92 (0,60-1,35)
1,16 (0,58-2,84)
30-39 1,83
(1,22-2,65)
1,46 (0,76-2,56)
40-44 1,52
(1,15-1,96) 1,67 (1,28-2,14)
0,99 (0,68-1,40)
1,47 (0,98-2,11)
1,55 (1,09-2,12)
1,27 (0,67-2,62)
4,31 (3,06-5,89)
1,90 (1,06-3,13)
45-49 1,54
(1,20-1,95) 1,21 (0,93-1,54)
1,58 (1,20-2,05)
1,04 (0,74-1,43)
1,35 (0,96-1,84)
1,56 (0,93-2,13)
2,58 (1,93-3,39)
1,65 (1,06-2,46)
50-54 1,63
(1,28-2,05) 1,64 (1,36-1,97)
2,87 (1,38-2,30)
1,68 (1,34-2,08)
1,70 (1,30-2,19)
1,77 (1,16-1,72)
4,07 (3,32-4,94)
2,82 (2,06-3,78)
55-59 1,67
(1,29-2,12) 1,87 (1,59-2,19)
2,87 (2,28-3,55)
1,46 (1,18-1,79)
1,46 (1,08-1,93)
1,33 (0,92-1,88)
4,03 (3,32-4,84)
2,79 (2,23-3,46)
60-64 2,29
(1,75-2,94) 1,65 (1,39-1,95)
2,04 (1,54-2,65)
1,72 (1,43-2,07)
1,41 (0,98-1,96)
1,77 (1,29-1,67)
3,26 (2,67-3,93)
2,93 (2,42-3,51)
65-69 2,37
(1,75-3,13) 1,72 (1,43-2,05)
1,70 (1,17-2,39)
1,57 (1,27-1,93)
1,54 (1,07-2,16)
1,58 (1,14-1,67)
3,01 (2,48-3,63)
2,01 (1,63-2,45)
70-74 2,05
(1,30-3,08) 1,63 (1,29-2,04)
2,21 (1,46-3,21)
1,68 (1,30-2,14)
1,85 (1,26-2,61)
1,35 (0,91-1,44)
2,35 (1,87-2,92)
1,75 (1,42-2,14)
75-79 1,74
(0,84-3,21) 1.30 (0,93-1,78)
2,19 (1,23-3,62)
1,49 (1,07-2,02)
1,65 (0,99-2,57)
1,58 (1,07-1,30)
1,53 (1,13-2,02)
1,40 (1,11-1,73)
80-84 1,76
(0,65-3,83) 0,87 (0,49-1,44)
2,02 (0,87-3,99)
0,97 (0,57-1,55)
1,75 (0,90-3,05)
1,34 (0,81-1,18)
1,40 (0,96-1,97)
1,27 (0,99-1,61)
>85 0,61 (0,02-3,42)
1,23 (0,74-1,92)
1,41 (0,29-4,13)
1,25 (0,74-1,98)
0,96 (0,26-2,45)
1.52 (0,95-2,30)
1,60 (1,07-2,30)
1,23 (0,99-1,52)
Table 4. Standardised incidence ratio (SIR) of low- and high-energy fractures by age groups in men and women with type 1 diabetes compared to those with type 2 diabetes across 20 years (1998-2017)
Table 5. Fractures were classified using the 10th edition of the International Classification of Diseases (ICD-10)
Low-energy fracture
High-energy fracture
Age groups Male
SIR(95%CL)
Female SIR(95% CL)
Male SIR(95% CL)
Female SIR(95% CL)
<30 0,90(0,84-0,97) 1,04(0,94-1,14) 1,11(1,02-1,21) 1,71(1,51-1,94)
30-34 1,09(0,95-1,25) 1,33(1.11-1,59) 2,12(1,77-1,66) 1,51(1,05-2,09)
35-39 1,15(1.00-1,30) 1,32(1.14-1,53) 1,25(1,04-1,64) 1,60(1,12-2,21)
40-44 1,45(1.30-1,61) 1,48(1.29-1.69) 1,26(1,06-1,40) 1,16(0,87-1,51)
45-49 1,81(1.64-1.99) 1,48(1.32-1.66) 1,56(1,34-1.34) 1,18(0,85-1,58)
50-54 1,98(1.82-2,16) 1,58(1.44-1.73) 1,48(1,26-1,10) 1,64(1,24-2,14)
55-59 2,12(1.94-2,31) 1,73(1.59-1.88) 1,61(1,37-1,09) 2,19(1,72-2,74)
60-64 2,15(1.96-2,35) 1,94(1.80-2,09) 1,35(1,08-0,90) 1,66(1,22-2,21)
65-69 2,03(1.82-2,25) 1,83(1.68-1.99) 1,55(1,20-0,93) 1,44(1,01-1,98)
70-74 2,00(1.76-2,26) 1,73(1.56-1.91) 1,30(0,88-0,93) 2,04(1,41-2,85)
75-79 1,72(1.46-2,02) 1,56(1.39-1.75) 1,20(0,66-0,92) 1,16(0,58-2,07)
80-84 1,32(1.04-1.64) 1,34(1.16-1.54) 1,51(0,65-1,01) 0,68(0,14-1,98)
>85 1,45(1,10-1,88) 1,37(1,20-1,56) 1,02(0,12-3,70) 2,28(0,92-4,71)
ICD-10 Non-
diabetics fractures
T1D fractures
T2D fractures
Hip S72.0, S72.1, S72.2 105609 1506 22617
Vertebral S22.0-1, S32.0, S32.7, M48.4-5 49991 779 7677
Proximal humerus S42.2 61287 1187 9240
Distal forearm S52.5, S52.6 180876 1958 10307
Hand, finger S62.0-7. 139887 1829 5155
Clavicula, scapula, rib, sternum
S22.2-4, S42.0-1, S42,7. 113006 1749 10614
Pelvis S32.3-8 28148 471 5162
Patella, tibia, fibula, malleol
S82 213432 3715 16133
Foot, toe, ankle S92 61668 1614 3920
Results
Both men and women with DM had an increased relative risk for distal forearm, proximal humerus, vertebral and hip fractures (Figure 1). Men and women with diabetes also had a higher relative risk of other fractures compared to people without DM (Supplementary table 1). Both among women and men, the distal forearm relative fracture risk was highest under the age of 50 (SIR 1.07–1.62).
Distal forearm fracture risk behaved differently compared to other fractures, as in men and women over the age of 55, the distal forearm fracture risk was lower among people with diabetes (Figure 1). For the proximal humerus, the relative fracture risk is particularly pronounced in the younger age groups with diabetes (Figure 1). The highest relative proximal humerus fracture risk was in the 30–34 age group among men with diabetes: the risk was over 3-fold higher compared to men in the control population (SIR 3.04; 95% CI 2.02–4.39), and nearly 4-fold higher in women (SIR 3.98;
95% CI 2.58–5.87). In both genders, the relative risk was higher in people with diabetes compared to people without diabetes, excluding those over the age of 85 (Figure 1). For vertebral fracture, the relative risk is pronounced in all ages and genders (Figure 1). The relative fracture risk was
especially pronounced in the younger groups. For hip fractures, the relative fracture risk was also particularly pronounced in the younger age groups with diabetes, and a similar occurrence is observed for the proximal humerus (Figure 1). The highest relative hip fracture risk was in men with diabetes between 30 and 54: the risk was over 2-fold higher compared to people without diabetes (SIR 2.09–2.72). In men over the age of 80, the relative fracture risk was lower in people with diabetes (SIR 0.77–0.96). Among women with diabetes, the highest relative hip fracture risk was under the age of 60: the risk was over 2-fold higher compared to people without diabetes (SIR 2.01–4.00). Finally, among women with diabetes aged over 85 years, relative hip fracture risk was almost the same (SIR 0.96; 95% CI 0.94–0.99).
Both men and women with DM had higher low- and high-energy relative fracture risk compared to people without diabetes (Figure 1). The highest relative low-energy fracture risk was found in men and women aged 30–44 years with DM, the risk was nearly 2-fold higher compared to the control population (SIR 1.74–2.08). With aging, the differences declined and the relative risk was lower in people with DM (Figure 1). High-energy fractures were more common among people under the age of 50 with diabetes and among people over the age of 50 without diabetes (Figure 1). In men over the age of 60, the relative risk for high-energy fractures became lower compared to people without diabetes (SIR 0.84–0.90). In women, the relative risk for high-energy fractures was also higher in people with diabetes under the age of 50 (SIR 1.08–1.48). In this group, people over the age of 50 had a lower risk for high-energy fractures compared to people without diabetes (SIR 0.79–0.90).
Men and women with type 1 diabetes had higher distal forearm, proximal humerus, vertebral and hip fracture relative risk in nearly all age groups compared to people with type 2 diabetes (Figure 2). Similar results are also observed for other fractures (Supplementary table 2). The relative risk
for distal forearm fractures was over 2-fold higher for type 1 diabetes compared to type 2 diabetes in men aged 60–74. Respectively, the relative risk was nearly 1.7-fold higher in women aged 60–74 and nearly 2-fold higher in women under the age of 40 with type 1 diabetes (Figure 2). In men with type 1 diabetes, the relative risk for proximal humerus fracture was pronounced in the older groups (Figure 2): the risk is over 2-fold higher compared to people with type 2 diabetes aged 70–84. The relative risk for vertebral fractures was highest in people aged 30–34 with type 1 diabetes in both genders (SIR 1.99–2.01) compared to people with type 2 diabetes. Men and women with type 1 diabetes had a higher relative hip fracture risk depending on age and sex compared to people with type 2 diabetes (Figure 2). At the age of 50–69, the relative hip fracture risk was over 3-fold higher in men with type 1 diabetes and over 2-fold higher in women with type 2 diabetes (Figure 2). With increasing age, the difference between type 1 and type 2 diabetes declined in both genders.
Men and women with type 1 diabetes had higher low-energy and high-energy fracture risk in nearly all age groups compared to people with type 2 diabetes (Figure 2). The highest low-energy relative risk was in men and women aged 60–64: the risk was nearly 2-fold higher compared to people with type 2 diabetes (SIR 2.15 and 1.94).
Discussion
In this study, we showed an increased risk of fractures overall, as well as hip, vertebral, proximal humerus, distal forearm and other fractures, among people with diabetes, depending on age and sex.
Also, we showed that people with type 1 diabetes had a higher relative fracture risk compared to people with type 2 diabetes in all fractures. The study included a population of all ages, and the age groups were divided into groups by sex and age (5-year age groups, under 30 years and over 85 years), which is new compared to previous studies.
Our study showed that, for a large proportion of fractures, the relative fracture risk associated with diabetes is noticeable, especially in adolescents. Under the age of 30, fractures almost invariably emerge in high-energy accidents such as car accidents. In young people, T1D is more common than T2D. According to a 2003 multinational retrospective survey by Cox et al., type 1 diabetics have an increased risk of being involved in a car accident. Study shows that clinical and treatment factors appear to increase risk, such as more frequent hypoglycaemia while driving, the method of insulin delivery, and infrequent self-testing before driving (23).
Our study shows significant differences between people with and without diabetes in terms of fractures. Previous studies suggest an association between diabetes and fractures (4,24). A retrospective study of 299,104 primary care patients showed that type 2 diabetes is related to an increased risk of fractures compared to people without diabetes (mean age 66 years), including fractures of the hip, spine, forearm and upper arm/shoulder(24). As in our study, the highest relative risks were found for upper arm and hip fractures. Stumpf et al. showed significant differences between people with type 1 diabetes and people without diabetes. Frequent fractures seen in that study were upper arm, sternum, thoracic spine, lower leg/ankle, foot and toe(4). In a case-control study, however, no greater risk of fracture was found in patients with diabetes than in matched control subjects(25). These studies are limited by small sample sizes and low event rates compared to our study.
We showed that people with type 1 diabetes had a higher fracture risk compared to people with type 2 diabetes. There are previous studies comparing fracture risk between type 1 diabetes and controls or type 2 diabetes and controls. The Nurses’ Health study showed that relative hip fracture risk in women with type 1 diabetes was 6.4, and with type 2 diabetes the relative risk was 2.2 compared to people without diabetes(19). In our study, we compared type 1 diabetes to type 2 diabetes, and the highest relative hip fracture risk was over 4-fold higher for men with type 1 diabetes and nearly 3- fold higher for women with type 1 diabetes, depending on age group, compared to people with type 2 diabetes. Previous studies showed an increased fracture risk and decreased bone mineral density in patients with complications of diabetes: retinopathy, neuropathy, nephropathy, and macro- and microangiopathy. There seems to be an association between diabetes and fractures, which is countered by certain factors in type 2 diabetes leading to a lower fracture risk than in patients with type 1 diabetes. One hypothesis could be that complications are responsible for the increased fracture risk through an increased risk of falls, and that the frequently increased body mass index in patients with type 2 diabetes tends to be protective and decrease the risk of fractures by increasing bone mineral density (26). Patients with type 1 diabetes, whose pathogenesis is due to decreased insulin secretion, have lower IGF-I levels than non-diabetics (27), and decreased bone mineral density (28). In contrast, people with type 2 diabetes are characterised by insulin resistance accompanied by obesity, and usually show higher serum levels of insulin and IGF-I than people with type 1 diabetes (27).These findings can also explain the relative fracture risk differences between people with type 1 and type 2 diabetes.
Our study showed that hip fractures were highly sensitive in both T1D and T2D compared to other fractures. The Iowa women’s health study reported that women with type 1 diabetes had a 12.25- fold higher hip fracture risk and women with type 2 diabetes had 1.70-fold higher hip fracture risk compared to people without diabetes. One hypothesis could be that falls related to diabetes may be the reason for the decreased bone mineral density in type 1 diabetes and a weakening of bone architecture through altered glycation of collagen, while retinopathy, neuropathy and obesity may be of greater significance in type 2 diabetes.
We also demonstrated a connection between diabetes and distal forearm fractures. Distal forearm fracture risk behaved differently compared to other fractures (Table 1), with the relative fracture risk being higher in men compared to women (Table 1). One hypothesis could be the impact of high body mass index, which can be a protective factor. A previous study shows that high BMI increased the risk of proximal humerus and ankle fractures, but lowered the risk of forearm fractures (29).
We also examined the incidence of low- and high-energy fractures in subjects with and without diabetes and with type 1 and type 2 diabetes. Low-energy fractures were more common among people under the age of 85 (SIR 1.07–2.45) with diabetes. This result is logical and expected, because most fractures are associated with the hip, and these fractures are generally due to low- energy accidents. Previous studies supported an association between diabetes and low-energy fractures: a meta-analysis showed that patients with type 2 diabetes had a higher risk of low-energy fractures compared to patients without diabetes (30). Another population-based study in Portugal consisted of 7675 subjects over 40 years old, reporting that patients with diabetes had a higher prevalence of self-reported low-energy fractures (31).
High-energy fractures were more common in people with diabetes, especially among younger people. The risk of having a high-energy fracture decreased with increasing age in people with DM.
This is possibly due to decreased activity in older people and reduced risk of high-energy fracture.
Under the age of 30, fractures almost invariably emerge in high-energy accidents, for example car accidents. There are no previous studies concerning high-energy fractures in people with diabetes.
One hypothesis could be that diabetes predisposes to high-energy trauma because of hypoglycaemia
while driving. Both men and women under the age of 30 with DM show a higher relative high- energy fracture risk compared to people without DM (SIR 1.43 and 1.48). The risk for high-energy fractures was 21% higher for people without diabetes in the 75–79 age group in women. For men, the risk was 16% higher in the 75–79 age group without diabetes. When separating type 1 and type 2 diabetes, we can notice that high-energy fracture risk is higher in people with type 1 diabetes compared to people with type 2 diabetes. Therefore, people with type 1 diabetes are a more fracture-sensitive population. One hypothesis is that there is an increased body mass index among people with type 2 diabetes compared to people with type 1 diabetes, which can be protective against fractures.
Strengths and limitations
There are no large studies comparing low- and high-energy fractures between genders in people with type 1 and type 2 diabetes. The strength of this study is the large number of study subjects and fractures. The association between diabetes and fractures has not previously been studied this comprehensively by age and gender. We were also able to study type 1 and type 2 diabetes
separately. Additionally, the study included people of all ages, with the strength of this study being the division of participants into groups by sex and age (5-year age groups, under 30 years and over 85 years).
A limitation of this study is the lack of information on the other risk factors for fractures. In
addition to diabetes, there are many other factors, such as blood glucose levels (HbA1C), degree of complications, body mass index, and use of tobacco, alcohol and medications. Further studies need to investigate fracture risk including these factors.
Conclusions
In conclusion, diabetes is associated with an increased risk of fractures, and the risk of low-energy fractures is higher among people with diabetes compared to non-diabetic populations. Respectively, high-energy fracture risk is higher among younger people with DM and older people without DM.
Comparing the risks between people with and without diabetes by age groups: the relative risk was higher in nearly every age group in hip, vertebral and proximal humerus fractures. Forearm fracture risk behaved differently compared to other fractures, because in men and women over the age of 55, the distal forearm fracture risk was higher in people without diabetes. In addition to the treatment of diabetes, it is important to keep in mind prevention of fractures among younger people, in whom diabetes can be a sign of increasing risk of fracture. Using bone density measurements it is possible to detect osteoporosis, but attention should also be paid to the prevention of falls and injuries among elderly people (32,33).
In summary, our study indicates that people with diabetes are at an increased risk of hip, vertebral, proximal humerus, distal forearm and other fractures depending on their age and sex. We showed significant differences in fracture risk between people with type 1 and type 2 diabetes. Type 1 DM was associated with higher relative fracture risk. Therefore, people with type 1 diabetes are the more fracture sensitive population. These findings highlight the need for studies including fracture prevention strategies in individuals with diabetes. Also, further studies are needed to investigate the incidence of fractures in younger subjects.
References
1. Janghorbani M, Van Dam RM, Willett WC, Hu FB. Systematic Review of Type 1 and Type 2 Diabetes Mellitus and Risk of Fracture. Am J Epidemiol [Internet]. 2007 Sep 1;166(5):495–505.
Available from: http://dx.doi.org/10.1093/aje/kwm106
2. Rubin MR. Type 2 diabetes and fractures: more information is needed. Endocrine [Internet].
2013;43(3):469–71. Available from: https://doi.org/10.1007/s12020-013-9901-6
3. Napoli N, Chandran M, Pierroz DD, Abrahamsen B, Schwartz A V, Ferrari SL, et al. Mechanisms of diabetes mellitus-induced bone fragility. Nat Rev Endocrinol [Internet]. 2017 Apr;13(4):208–19.
Available from: http://dx.doi.org/10.1038/nrendo.2016.153
4. Stumpf U, Hadji P, van den Boom L, Böcker W, Kostev K. Incidence of fractures in patients with type 1 diabetes mellitus—a retrospective study with 4420 patients. Osteoporos Int [Internet].
2020;31(7):1315–22. Available from: https://doi.org/10.1007/s00198-020-05344-w
5. Vilaca T, Schini M, Harnan S, Sutton A, Poku E, Allen IE, et al. The risk of hip and non-vertebral fractures in type 1 and type 2 diabetes: A systematic review and meta-analysis update. Bone [Internet]. 2020;137:115457. Available from:
http://www.sciencedirect.com/science/article/pii/S8756328220302374
6. Koromani F, Oei L, Shevroja E, Trajanoska K, Schoufour J, Muka T, et al. Vertebral Fractures in Individuals With Type 2 Diabetes: More Than Skeletal Complications Alone. Diabetes Care [Internet]. 2019/10/28. 2020 Jan;43(1):137–44. Available from:
https://pubmed.ncbi.nlm.nih.gov/31658976
7. Vestergaard P, Rejnmark L, Mosekilde L. Relative fracture risk in patients with diabetes mellitus, and the impact of insulin and oral antidiabetic medication on relative fracture risk. Diabetologia
[Internet]. 2005;48(7):1292–9. Available from: https://doi.org/10.1007/s00125-005-1786-3
8. Jensen MH, Vestergaard P. Hypoglycaemia and type 1 diabetes are associated with an increased risk of fractures. Osteoporos Int [Internet]. 2019;30(8):1663–70. Available from:
https://doi.org/10.1007/s00198-019-05014-6
9. Napoli N, Strotmeyer ES, Ensrud KE, Sellmeyer DE, Bauer DC, Hoffman AR, et al. Fracture risk in diabetic elderly men: the MrOS study. Diabetologia [Internet]. 2014;57(10):2057–65. Available from:
https://doi.org/10.1007/s00125-014-3289-6
10. Huang H-L, Pan C-C, Hsiao Y-F, Chen M-C, Kung C-Y, Kung P-T, et al. Associations of body mass
index and diabetes with hip fracture risk: a nationwide cohort study. BMC Public Health. 2018 Nov;18(1):1325.
11. Rasmussen NH, Dal J, de Vries F, van den Bergh JP, Jensen MH, Vestergaard P. Diabetes and fractures: new evidence of atypical femoral fractures? Osteoporos Int [Internet]. 2020;31(3):447–55.
Available from: https://doi.org/10.1007/s00198-019-05224-y
12. Ahmed LA, Joakimsen RM, Berntsen GK, Fønnebø V, Schirmer H. Diabetes mellitus and the risk of non-vertebral fractures: the Tromsø study. Osteoporos Int [Internet]. 2006;17(4):495–500. Available from: https://doi.org/10.1007/s00198-005-0013-x
13. Schwartz A V, Sellmeyer DE, Ensrud KE, Cauley JA, Tabor HK, Schreiner PJ, et al. Older Women with Diabetes Have an Increased Risk of Fracture: A Prospective Study. J Clin Endocrinol Metab [Internet]. 2001 Jan 1;86(1):32–8. Available from: http://dx.doi.org/10.1210/jcem.86.1.7139 14. Thong EP, Milat F, Enticott JC, Joham AE, Ebeling PR, Mishra GD, et al. The diabetes-fracture
association in women with type 1 and type 2 diabetes is partially mediated by falls: a 15-year longitudinal study. Osteoporos Int [Internet]. 2021; Available from: https://doi.org/10.1007/s00198- 020-05771-9
15. Vestergaard P. Discrepancies in bone mineral density and fracture risk in patients with type 1 and type 2 diabetes---a meta-analysis. Osteoporos Int [Internet]. 2007;18(4):427–44. Available from:
https://doi.org/10.1007/s00198-006-0253-4
16. Nicodemus KK, Folsom AR. Type 1 and Type 2 Diabetes and Incident Hip Fractures in Postmenopausal Women. Diabetes Care. 2001 Jul;24(7):1192 LP – 1197.
17. de Liefde II, van der Klift M, de Laet CEDH, van Daele PLA, Hofman A, Pols HAP. Bone mineral density and fracture risk in type-2 diabetes mellitus: the Rotterdam Study. Osteoporos Int [Internet].
2005;16(12):1713–20. Available from: https://doi.org/10.1007/s00198-005-1909-1
18. Bonds DE, Larson JC, Schwartz A V, Strotmeyer ES, Robbins J, Rodriguez BL, et al. Risk of Fracture in Women with Type 2 Diabetes: the Women’s Health Initiative Observational Study. J Clin Endocrinol Metab [Internet]. 2006 Sep 1;91(9):3404–10. Available from:
http://dx.doi.org/10.1210/jc.2006-0614
19. Janghorbani M, Feskanich D, Willett WC, Hu F. Prospective Study of Diabetes and Risk of Hip Fracture. Diabetes Care [Internet]. 2006 Jun 26;29(7):1573 LP – 1578. Available from:
http://care.diabetesjournals.org/content/29/7/1573.abstract
20. Sund R, Koski S. Diabeteksen ja sen lisäsairauksien esiintyvyyden ja ilmaantuvuuden
rekisteriperusteinen mittaaminen. 2009;13.
21. Sund R. Utilization of routinely collected administrative data in monitoring the incidence of aging dependent hip fracture. Epidemiol Perspect Innov [Internet]. 2007 Jun 7;4:2. Available from:
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1899499/
22. Sotkanet [Internet]. Available from: https://sotkanet.fi/sotkanet/fi/index
23. Cox DJ, Penberthy JK, Zrebiec J, Weinger K, Aikens JE, Frier B, et al. Diabetes and Driving Mishaps. Diabetes Care [Internet]. 2003 Aug 1;26(8):2329 LP – 2334. Available from:
http://care.diabetesjournals.org/content/26/8/2329.abstract
24. Rathmann W, Kostev K. Fracture risk in patients with newly diagnosed type 2 diabetes: a
retrospective database analysis in primary care. J Diabetes Complications [Internet]. 2015;29(6):766–
70. Available from: http://www.sciencedirect.com/science/article/pii/S1056872715001981
25. Heath III, MD H, Melton III, MD LJ, Chu MS C-P. Diabetes Mellitus and Risk of Skeletal Fracture.
N Engl J Med [Internet]. 1980 Sep 4;303(10):567–70. Available from:
http://ezproxy.uef.fi:2048/login?url=https://www.proquest.com/scholarly-journals/diabetes-mellitus- risk-skeletal-fracture/docview/1869576749/se-2?accountid=11739
26. De Laet C, Kanis JA, Odén A, Johanson H, Johnell O, Delmas P, et al. Body mass index as a
predictor of fracture risk: A meta-analysis. Osteoporos Int [Internet]. 2005;16(11):1330–8. Available from: https://doi.org/10.1007/s00198-005-1863-y
27. Weinstock RS, Goland RS, Shane E, Clemens TL, Lindsay R, Bilezikian JP. Bone mineral density in women with type II diabetes mellitus. J Bone Miner Res [Internet]. 1989 Feb 1;4(1):97–101.
Available from: https://doi.org/10.1002/jbmr.5650040114
28. Nisbeth U, Lindh E, Ljunghall S, Backman U, Fellström B. INCREASED FRACTURE RATE IN DIABETES MELLITUS AND FEMALES AFTER RENAL TRANSPLANTATION1.
Transplantation [Internet]. 1999;67(9). Available from:
https://journals.lww.com/transplantjournal/Fulltext/1999/05150/INCREASED_FRACTURE_RATE_
IN_DIABETES_MELLITUS_AND.4.aspx
29. Holmberg AH, Johnell O, Nilsson PM, Nilsson J, Berglund G, Åkesson K. Risk factors for fragility fracture in middle age. A prospective population-based study of 33,000 men and women. Osteoporos Int [Internet]. 2006;17(7):1065–77. Available from: https://doi.org/10.1007/s00198-006-0137-7 30. Jia P, Bao L, Chen H, Yuan J, Liu W, Feng F, et al. Risk of low-energy fracture in type 2 diabetes
patients: a meta-analysis of observational studies. Osteoporos Int [Internet]. 2017;28(11):3113–21.
Available from: https://doi.org/10.1007/s00198-017-4183-0
31. Furtado S, Rodrigues A, Dias S, Branco JC, Canhão H. Self-reported low-energy fractures and associated risk factors in people with diabetes: A national population-based study. Diabetes Res Clin Pract [Internet]. 2019;147:93–101. Available from:
http://www.sciencedirect.com/science/article/pii/S0168822718308386
32. Gonnelli S, Caffarelli C, Giordano N, Nuti R. The prevention of fragility fractures in diabetic patients. Aging Clin Exp Res [Internet]. 2015;27(2):115–24. Available from:
https://doi.org/10.1007/s40520-014-0258-3
33. Rasmussen NH, Dal J. Falls and Fractures in Diabetes---More than Bone Fragility. Curr Osteoporos Rep. 2019;17(3):147–56.
